Enhancing Elementary Students’ Learning Interest Through AI-Driven Canva Learning Tools: Evidence from a Pre-Experimental Study

Abstract

This study investigates the effectiveness of an AI-driven Canva learning tool in enhancing elementary students’ learning interest within a pre-experimental pretest–posttest design. Twenty-five Grade 5 students from SD Baitul Amien Jember participated in a two-week intervention consisting of six sessions that integrated adaptive learning systems, personalized feedback, visual learning design, and gamified tasks. Learning interest was measured using a validated 20-item questionnaire covering attention, curiosity, persistence, enjoyment, engagement, and relevance. Descriptive statistics show a substantial improvement from the pretest (M = 44.08, SD = 8.37) to the posttest (M = 81.84, SD = 6.26). The paired sample t-test confirmed a highly significant effect, t(24) = –15.611, p < .001, with a very large effect size (Cohen’s d = 3.12). These findings demonstrate that AI-driven Canva learning tools meaningfully enhance students’ learning interest by providing adaptive, visually rich, and personalized learning experiences. The study contributes empirical evidence to the growing field of AI-supported learning in primary education and highlights the need for teacher digital-pedagogy readiness and adequate technological infrastructure to ensure effective AI integration.